Training in progress, step 6500, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 536223056
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8da83f3c30b9473fef2b931e6b47e4814c76e805b02501c93641aed6bc786ead
|
| 3 |
size 536223056
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1072594443
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d8c7391e5803dc14420bae3b5326bbd52abb5236b17e67147b31348d199ebeef
|
| 3 |
size 1072594443
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0717c8780efa444a6d80d462b725b32f107f9a3c24550aaaa04a7d27cefba76b
|
| 3 |
size 1465
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 1.
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -5408,6 +5408,456 @@
|
|
| 5408 |
"mean_token_accuracy": 0.8181872367858887,
|
| 5409 |
"num_tokens": 6657605.0,
|
| 5410 |
"step": 6000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5411 |
}
|
| 5412 |
],
|
| 5413 |
"logging_steps": 10,
|
|
@@ -5427,7 +5877,7 @@
|
|
| 5427 |
"attributes": {}
|
| 5428 |
}
|
| 5429 |
},
|
| 5430 |
-
"total_flos":
|
| 5431 |
"train_batch_size": 8,
|
| 5432 |
"trial_name": null,
|
| 5433 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.309691718718517,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 6500,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 5408 |
"mean_token_accuracy": 0.8181872367858887,
|
| 5409 |
"num_tokens": 6657605.0,
|
| 5410 |
"step": 6000
|
| 5411 |
+
},
|
| 5412 |
+
{
|
| 5413 |
+
"epoch": 1.2109611122305057,
|
| 5414 |
+
"grad_norm": 16.125,
|
| 5415 |
+
"learning_rate": 1.1928269192020955e-05,
|
| 5416 |
+
"loss": 0.8571,
|
| 5417 |
+
"mean_token_accuracy": 0.7833750724792481,
|
| 5418 |
+
"num_tokens": 6667905.0,
|
| 5419 |
+
"step": 6010
|
| 5420 |
+
},
|
| 5421 |
+
{
|
| 5422 |
+
"epoch": 1.2129760225669957,
|
| 5423 |
+
"grad_norm": 9.8125,
|
| 5424 |
+
"learning_rate": 1.1914836456444356e-05,
|
| 5425 |
+
"loss": 0.7898,
|
| 5426 |
+
"mean_token_accuracy": 0.7975350022315979,
|
| 5427 |
+
"num_tokens": 6678888.0,
|
| 5428 |
+
"step": 6020
|
| 5429 |
+
},
|
| 5430 |
+
{
|
| 5431 |
+
"epoch": 1.2149909329034858,
|
| 5432 |
+
"grad_norm": 13.5,
|
| 5433 |
+
"learning_rate": 1.1901403720867755e-05,
|
| 5434 |
+
"loss": 0.8104,
|
| 5435 |
+
"mean_token_accuracy": 0.7968179106712341,
|
| 5436 |
+
"num_tokens": 6690767.0,
|
| 5437 |
+
"step": 6030
|
| 5438 |
+
},
|
| 5439 |
+
{
|
| 5440 |
+
"epoch": 1.2170058432399757,
|
| 5441 |
+
"grad_norm": 12.9375,
|
| 5442 |
+
"learning_rate": 1.1887970985291155e-05,
|
| 5443 |
+
"loss": 0.8847,
|
| 5444 |
+
"mean_token_accuracy": 0.7826601445674897,
|
| 5445 |
+
"num_tokens": 6701963.0,
|
| 5446 |
+
"step": 6040
|
| 5447 |
+
},
|
| 5448 |
+
{
|
| 5449 |
+
"epoch": 1.2190207535764659,
|
| 5450 |
+
"grad_norm": 10.375,
|
| 5451 |
+
"learning_rate": 1.1874538249714556e-05,
|
| 5452 |
+
"loss": 0.7929,
|
| 5453 |
+
"mean_token_accuracy": 0.795721584558487,
|
| 5454 |
+
"num_tokens": 6712765.0,
|
| 5455 |
+
"step": 6050
|
| 5456 |
+
},
|
| 5457 |
+
{
|
| 5458 |
+
"epoch": 1.2210356639129558,
|
| 5459 |
+
"grad_norm": 9.5625,
|
| 5460 |
+
"learning_rate": 1.1861105514137956e-05,
|
| 5461 |
+
"loss": 0.8396,
|
| 5462 |
+
"mean_token_accuracy": 0.7961892068386078,
|
| 5463 |
+
"num_tokens": 6723997.0,
|
| 5464 |
+
"step": 6060
|
| 5465 |
+
},
|
| 5466 |
+
{
|
| 5467 |
+
"epoch": 1.223050574249446,
|
| 5468 |
+
"grad_norm": 10.4375,
|
| 5469 |
+
"learning_rate": 1.1847672778561354e-05,
|
| 5470 |
+
"loss": 0.7774,
|
| 5471 |
+
"mean_token_accuracy": 0.8090421617031097,
|
| 5472 |
+
"num_tokens": 6734959.0,
|
| 5473 |
+
"step": 6070
|
| 5474 |
+
},
|
| 5475 |
+
{
|
| 5476 |
+
"epoch": 1.225065484585936,
|
| 5477 |
+
"grad_norm": 13.75,
|
| 5478 |
+
"learning_rate": 1.1834240042984754e-05,
|
| 5479 |
+
"loss": 0.8442,
|
| 5480 |
+
"mean_token_accuracy": 0.7920287191867829,
|
| 5481 |
+
"num_tokens": 6745125.0,
|
| 5482 |
+
"step": 6080
|
| 5483 |
+
},
|
| 5484 |
+
{
|
| 5485 |
+
"epoch": 1.227080394922426,
|
| 5486 |
+
"grad_norm": 12.375,
|
| 5487 |
+
"learning_rate": 1.1820807307408155e-05,
|
| 5488 |
+
"loss": 0.7339,
|
| 5489 |
+
"mean_token_accuracy": 0.8152937352657318,
|
| 5490 |
+
"num_tokens": 6755578.0,
|
| 5491 |
+
"step": 6090
|
| 5492 |
+
},
|
| 5493 |
+
{
|
| 5494 |
+
"epoch": 1.229095305258916,
|
| 5495 |
+
"grad_norm": 10.5,
|
| 5496 |
+
"learning_rate": 1.1807374571831553e-05,
|
| 5497 |
+
"loss": 0.7799,
|
| 5498 |
+
"mean_token_accuracy": 0.8055865943431855,
|
| 5499 |
+
"num_tokens": 6766161.0,
|
| 5500 |
+
"step": 6100
|
| 5501 |
+
},
|
| 5502 |
+
{
|
| 5503 |
+
"epoch": 1.2311102155954061,
|
| 5504 |
+
"grad_norm": 12.5,
|
| 5505 |
+
"learning_rate": 1.1793941836254954e-05,
|
| 5506 |
+
"loss": 0.8072,
|
| 5507 |
+
"mean_token_accuracy": 0.7963293552398681,
|
| 5508 |
+
"num_tokens": 6776660.0,
|
| 5509 |
+
"step": 6110
|
| 5510 |
+
},
|
| 5511 |
+
{
|
| 5512 |
+
"epoch": 1.233125125931896,
|
| 5513 |
+
"grad_norm": 14.0,
|
| 5514 |
+
"learning_rate": 1.1780509100678354e-05,
|
| 5515 |
+
"loss": 0.882,
|
| 5516 |
+
"mean_token_accuracy": 0.7818022012710572,
|
| 5517 |
+
"num_tokens": 6787702.0,
|
| 5518 |
+
"step": 6120
|
| 5519 |
+
},
|
| 5520 |
+
{
|
| 5521 |
+
"epoch": 1.235140036268386,
|
| 5522 |
+
"grad_norm": 11.1875,
|
| 5523 |
+
"learning_rate": 1.1767076365101755e-05,
|
| 5524 |
+
"loss": 0.8014,
|
| 5525 |
+
"mean_token_accuracy": 0.8063171863555908,
|
| 5526 |
+
"num_tokens": 6798904.0,
|
| 5527 |
+
"step": 6130
|
| 5528 |
+
},
|
| 5529 |
+
{
|
| 5530 |
+
"epoch": 1.2371549466048761,
|
| 5531 |
+
"grad_norm": 13.75,
|
| 5532 |
+
"learning_rate": 1.1753643629525154e-05,
|
| 5533 |
+
"loss": 0.8453,
|
| 5534 |
+
"mean_token_accuracy": 0.7982756316661834,
|
| 5535 |
+
"num_tokens": 6808990.0,
|
| 5536 |
+
"step": 6140
|
| 5537 |
+
},
|
| 5538 |
+
{
|
| 5539 |
+
"epoch": 1.239169856941366,
|
| 5540 |
+
"grad_norm": 10.8125,
|
| 5541 |
+
"learning_rate": 1.1740210893948554e-05,
|
| 5542 |
+
"loss": 0.7952,
|
| 5543 |
+
"mean_token_accuracy": 0.8004013955593109,
|
| 5544 |
+
"num_tokens": 6818726.0,
|
| 5545 |
+
"step": 6150
|
| 5546 |
+
},
|
| 5547 |
+
{
|
| 5548 |
+
"epoch": 1.2411847672778562,
|
| 5549 |
+
"grad_norm": 9.3125,
|
| 5550 |
+
"learning_rate": 1.1726778158371955e-05,
|
| 5551 |
+
"loss": 0.8466,
|
| 5552 |
+
"mean_token_accuracy": 0.794275438785553,
|
| 5553 |
+
"num_tokens": 6830189.0,
|
| 5554 |
+
"step": 6160
|
| 5555 |
+
},
|
| 5556 |
+
{
|
| 5557 |
+
"epoch": 1.2431996776143461,
|
| 5558 |
+
"grad_norm": 11.8125,
|
| 5559 |
+
"learning_rate": 1.1713345422795352e-05,
|
| 5560 |
+
"loss": 0.809,
|
| 5561 |
+
"mean_token_accuracy": 0.7935117900371551,
|
| 5562 |
+
"num_tokens": 6840564.0,
|
| 5563 |
+
"step": 6170
|
| 5564 |
+
},
|
| 5565 |
+
{
|
| 5566 |
+
"epoch": 1.2452145879508363,
|
| 5567 |
+
"grad_norm": 10.1875,
|
| 5568 |
+
"learning_rate": 1.1699912687218753e-05,
|
| 5569 |
+
"loss": 0.8684,
|
| 5570 |
+
"mean_token_accuracy": 0.7853965878486633,
|
| 5571 |
+
"num_tokens": 6851057.0,
|
| 5572 |
+
"step": 6180
|
| 5573 |
+
},
|
| 5574 |
+
{
|
| 5575 |
+
"epoch": 1.2472294982873262,
|
| 5576 |
+
"grad_norm": 12.3125,
|
| 5577 |
+
"learning_rate": 1.1686479951642153e-05,
|
| 5578 |
+
"loss": 0.7914,
|
| 5579 |
+
"mean_token_accuracy": 0.7986515760421753,
|
| 5580 |
+
"num_tokens": 6863098.0,
|
| 5581 |
+
"step": 6190
|
| 5582 |
+
},
|
| 5583 |
+
{
|
| 5584 |
+
"epoch": 1.2492444086238161,
|
| 5585 |
+
"grad_norm": 11.1875,
|
| 5586 |
+
"learning_rate": 1.1673047216065552e-05,
|
| 5587 |
+
"loss": 0.7723,
|
| 5588 |
+
"mean_token_accuracy": 0.8050061583518981,
|
| 5589 |
+
"num_tokens": 6873361.0,
|
| 5590 |
+
"step": 6200
|
| 5591 |
+
},
|
| 5592 |
+
{
|
| 5593 |
+
"epoch": 1.2512593189603063,
|
| 5594 |
+
"grad_norm": 10.6875,
|
| 5595 |
+
"learning_rate": 1.1659614480488952e-05,
|
| 5596 |
+
"loss": 0.7892,
|
| 5597 |
+
"mean_token_accuracy": 0.8081447362899781,
|
| 5598 |
+
"num_tokens": 6885561.0,
|
| 5599 |
+
"step": 6210
|
| 5600 |
+
},
|
| 5601 |
+
{
|
| 5602 |
+
"epoch": 1.2532742292967962,
|
| 5603 |
+
"grad_norm": 11.5625,
|
| 5604 |
+
"learning_rate": 1.1646181744912353e-05,
|
| 5605 |
+
"loss": 0.8335,
|
| 5606 |
+
"mean_token_accuracy": 0.7930399179458618,
|
| 5607 |
+
"num_tokens": 6896678.0,
|
| 5608 |
+
"step": 6220
|
| 5609 |
+
},
|
| 5610 |
+
{
|
| 5611 |
+
"epoch": 1.2552891396332864,
|
| 5612 |
+
"grad_norm": 12.5,
|
| 5613 |
+
"learning_rate": 1.1632749009335754e-05,
|
| 5614 |
+
"loss": 0.8161,
|
| 5615 |
+
"mean_token_accuracy": 0.8012160181999206,
|
| 5616 |
+
"num_tokens": 6906436.0,
|
| 5617 |
+
"step": 6230
|
| 5618 |
+
},
|
| 5619 |
+
{
|
| 5620 |
+
"epoch": 1.2573040499697763,
|
| 5621 |
+
"grad_norm": 14.5625,
|
| 5622 |
+
"learning_rate": 1.161931627375915e-05,
|
| 5623 |
+
"loss": 0.8408,
|
| 5624 |
+
"mean_token_accuracy": 0.7945603370666504,
|
| 5625 |
+
"num_tokens": 6916620.0,
|
| 5626 |
+
"step": 6240
|
| 5627 |
+
},
|
| 5628 |
+
{
|
| 5629 |
+
"epoch": 1.2593189603062664,
|
| 5630 |
+
"grad_norm": 16.25,
|
| 5631 |
+
"learning_rate": 1.1605883538182551e-05,
|
| 5632 |
+
"loss": 0.8039,
|
| 5633 |
+
"mean_token_accuracy": 0.801008677482605,
|
| 5634 |
+
"num_tokens": 6928094.0,
|
| 5635 |
+
"step": 6250
|
| 5636 |
+
},
|
| 5637 |
+
{
|
| 5638 |
+
"epoch": 1.2613338706427564,
|
| 5639 |
+
"grad_norm": 11.0625,
|
| 5640 |
+
"learning_rate": 1.1592450802605952e-05,
|
| 5641 |
+
"loss": 0.7906,
|
| 5642 |
+
"mean_token_accuracy": 0.8035953044891357,
|
| 5643 |
+
"num_tokens": 6937852.0,
|
| 5644 |
+
"step": 6260
|
| 5645 |
+
},
|
| 5646 |
+
{
|
| 5647 |
+
"epoch": 1.2633487809792463,
|
| 5648 |
+
"grad_norm": 13.0,
|
| 5649 |
+
"learning_rate": 1.157901806702935e-05,
|
| 5650 |
+
"loss": 0.8865,
|
| 5651 |
+
"mean_token_accuracy": 0.7758103013038635,
|
| 5652 |
+
"num_tokens": 6948566.0,
|
| 5653 |
+
"step": 6270
|
| 5654 |
+
},
|
| 5655 |
+
{
|
| 5656 |
+
"epoch": 1.2653636913157364,
|
| 5657 |
+
"grad_norm": 14.125,
|
| 5658 |
+
"learning_rate": 1.1565585331452751e-05,
|
| 5659 |
+
"loss": 0.9311,
|
| 5660 |
+
"mean_token_accuracy": 0.7705212533473969,
|
| 5661 |
+
"num_tokens": 6959264.0,
|
| 5662 |
+
"step": 6280
|
| 5663 |
+
},
|
| 5664 |
+
{
|
| 5665 |
+
"epoch": 1.2673786016522266,
|
| 5666 |
+
"grad_norm": 11.3125,
|
| 5667 |
+
"learning_rate": 1.1552152595876152e-05,
|
| 5668 |
+
"loss": 0.8076,
|
| 5669 |
+
"mean_token_accuracy": 0.7959451377391815,
|
| 5670 |
+
"num_tokens": 6970706.0,
|
| 5671 |
+
"step": 6290
|
| 5672 |
+
},
|
| 5673 |
+
{
|
| 5674 |
+
"epoch": 1.2693935119887165,
|
| 5675 |
+
"grad_norm": 12.9375,
|
| 5676 |
+
"learning_rate": 1.1538719860299552e-05,
|
| 5677 |
+
"loss": 0.8435,
|
| 5678 |
+
"mean_token_accuracy": 0.7901066780090332,
|
| 5679 |
+
"num_tokens": 6982005.0,
|
| 5680 |
+
"step": 6300
|
| 5681 |
+
},
|
| 5682 |
+
{
|
| 5683 |
+
"epoch": 1.2714084223252065,
|
| 5684 |
+
"grad_norm": 15.875,
|
| 5685 |
+
"learning_rate": 1.1525287124722951e-05,
|
| 5686 |
+
"loss": 0.7781,
|
| 5687 |
+
"mean_token_accuracy": 0.8100695073604584,
|
| 5688 |
+
"num_tokens": 6993228.0,
|
| 5689 |
+
"step": 6310
|
| 5690 |
+
},
|
| 5691 |
+
{
|
| 5692 |
+
"epoch": 1.2734233326616966,
|
| 5693 |
+
"grad_norm": 11.8125,
|
| 5694 |
+
"learning_rate": 1.1511854389146352e-05,
|
| 5695 |
+
"loss": 0.7771,
|
| 5696 |
+
"mean_token_accuracy": 0.7994659662246704,
|
| 5697 |
+
"num_tokens": 7003855.0,
|
| 5698 |
+
"step": 6320
|
| 5699 |
+
},
|
| 5700 |
+
{
|
| 5701 |
+
"epoch": 1.2754382429981865,
|
| 5702 |
+
"grad_norm": 11.5625,
|
| 5703 |
+
"learning_rate": 1.1498421653569752e-05,
|
| 5704 |
+
"loss": 0.8883,
|
| 5705 |
+
"mean_token_accuracy": 0.7779460906982422,
|
| 5706 |
+
"num_tokens": 7016389.0,
|
| 5707 |
+
"step": 6330
|
| 5708 |
+
},
|
| 5709 |
+
{
|
| 5710 |
+
"epoch": 1.2774531533346767,
|
| 5711 |
+
"grad_norm": 10.6875,
|
| 5712 |
+
"learning_rate": 1.148498891799315e-05,
|
| 5713 |
+
"loss": 0.7674,
|
| 5714 |
+
"mean_token_accuracy": 0.8048594057559967,
|
| 5715 |
+
"num_tokens": 7027229.0,
|
| 5716 |
+
"step": 6340
|
| 5717 |
+
},
|
| 5718 |
+
{
|
| 5719 |
+
"epoch": 1.2794680636711666,
|
| 5720 |
+
"grad_norm": 13.0,
|
| 5721 |
+
"learning_rate": 1.147155618241655e-05,
|
| 5722 |
+
"loss": 0.8556,
|
| 5723 |
+
"mean_token_accuracy": 0.7895182788372039,
|
| 5724 |
+
"num_tokens": 7038636.0,
|
| 5725 |
+
"step": 6350
|
| 5726 |
+
},
|
| 5727 |
+
{
|
| 5728 |
+
"epoch": 1.2814829740076568,
|
| 5729 |
+
"grad_norm": 10.75,
|
| 5730 |
+
"learning_rate": 1.145812344683995e-05,
|
| 5731 |
+
"loss": 0.8673,
|
| 5732 |
+
"mean_token_accuracy": 0.7855879724025726,
|
| 5733 |
+
"num_tokens": 7049602.0,
|
| 5734 |
+
"step": 6360
|
| 5735 |
+
},
|
| 5736 |
+
{
|
| 5737 |
+
"epoch": 1.2834978843441467,
|
| 5738 |
+
"grad_norm": 11.0625,
|
| 5739 |
+
"learning_rate": 1.1444690711263349e-05,
|
| 5740 |
+
"loss": 0.8281,
|
| 5741 |
+
"mean_token_accuracy": 0.798017168045044,
|
| 5742 |
+
"num_tokens": 7060783.0,
|
| 5743 |
+
"step": 6370
|
| 5744 |
+
},
|
| 5745 |
+
{
|
| 5746 |
+
"epoch": 1.2855127946806366,
|
| 5747 |
+
"grad_norm": 10.5,
|
| 5748 |
+
"learning_rate": 1.143125797568675e-05,
|
| 5749 |
+
"loss": 0.781,
|
| 5750 |
+
"mean_token_accuracy": 0.8092272758483887,
|
| 5751 |
+
"num_tokens": 7072749.0,
|
| 5752 |
+
"step": 6380
|
| 5753 |
+
},
|
| 5754 |
+
{
|
| 5755 |
+
"epoch": 1.2875277050171268,
|
| 5756 |
+
"grad_norm": 9.75,
|
| 5757 |
+
"learning_rate": 1.141782524011015e-05,
|
| 5758 |
+
"loss": 0.8664,
|
| 5759 |
+
"mean_token_accuracy": 0.7885208010673523,
|
| 5760 |
+
"num_tokens": 7085190.0,
|
| 5761 |
+
"step": 6390
|
| 5762 |
+
},
|
| 5763 |
+
{
|
| 5764 |
+
"epoch": 1.2895426153536167,
|
| 5765 |
+
"grad_norm": 10.625,
|
| 5766 |
+
"learning_rate": 1.140439250453355e-05,
|
| 5767 |
+
"loss": 0.7861,
|
| 5768 |
+
"mean_token_accuracy": 0.80192711353302,
|
| 5769 |
+
"num_tokens": 7095259.0,
|
| 5770 |
+
"step": 6400
|
| 5771 |
+
},
|
| 5772 |
+
{
|
| 5773 |
+
"epoch": 1.2915575256901068,
|
| 5774 |
+
"grad_norm": 10.5625,
|
| 5775 |
+
"learning_rate": 1.1390959768956948e-05,
|
| 5776 |
+
"loss": 0.7456,
|
| 5777 |
+
"mean_token_accuracy": 0.8084029912948608,
|
| 5778 |
+
"num_tokens": 7104840.0,
|
| 5779 |
+
"step": 6410
|
| 5780 |
+
},
|
| 5781 |
+
{
|
| 5782 |
+
"epoch": 1.2935724360265968,
|
| 5783 |
+
"grad_norm": 9.25,
|
| 5784 |
+
"learning_rate": 1.1377527033380348e-05,
|
| 5785 |
+
"loss": 0.9101,
|
| 5786 |
+
"mean_token_accuracy": 0.7775183081626892,
|
| 5787 |
+
"num_tokens": 7116704.0,
|
| 5788 |
+
"step": 6420
|
| 5789 |
+
},
|
| 5790 |
+
{
|
| 5791 |
+
"epoch": 1.295587346363087,
|
| 5792 |
+
"grad_norm": 10.5625,
|
| 5793 |
+
"learning_rate": 1.1364094297803749e-05,
|
| 5794 |
+
"loss": 0.8154,
|
| 5795 |
+
"mean_token_accuracy": 0.7952620327472687,
|
| 5796 |
+
"num_tokens": 7128360.0,
|
| 5797 |
+
"step": 6430
|
| 5798 |
+
},
|
| 5799 |
+
{
|
| 5800 |
+
"epoch": 1.2976022566995769,
|
| 5801 |
+
"grad_norm": 10.0625,
|
| 5802 |
+
"learning_rate": 1.1350661562227148e-05,
|
| 5803 |
+
"loss": 0.7922,
|
| 5804 |
+
"mean_token_accuracy": 0.8016098260879516,
|
| 5805 |
+
"num_tokens": 7139177.0,
|
| 5806 |
+
"step": 6440
|
| 5807 |
+
},
|
| 5808 |
+
{
|
| 5809 |
+
"epoch": 1.2996171670360668,
|
| 5810 |
+
"grad_norm": 13.0625,
|
| 5811 |
+
"learning_rate": 1.1337228826650548e-05,
|
| 5812 |
+
"loss": 0.9353,
|
| 5813 |
+
"mean_token_accuracy": 0.7753040254116058,
|
| 5814 |
+
"num_tokens": 7149827.0,
|
| 5815 |
+
"step": 6450
|
| 5816 |
+
},
|
| 5817 |
+
{
|
| 5818 |
+
"epoch": 1.301632077372557,
|
| 5819 |
+
"grad_norm": 9.4375,
|
| 5820 |
+
"learning_rate": 1.1323796091073949e-05,
|
| 5821 |
+
"loss": 0.8437,
|
| 5822 |
+
"mean_token_accuracy": 0.7929971814155579,
|
| 5823 |
+
"num_tokens": 7162136.0,
|
| 5824 |
+
"step": 6460
|
| 5825 |
+
},
|
| 5826 |
+
{
|
| 5827 |
+
"epoch": 1.303646987709047,
|
| 5828 |
+
"grad_norm": 10.875,
|
| 5829 |
+
"learning_rate": 1.1310363355497348e-05,
|
| 5830 |
+
"loss": 0.9565,
|
| 5831 |
+
"mean_token_accuracy": 0.7702773094177247,
|
| 5832 |
+
"num_tokens": 7173369.0,
|
| 5833 |
+
"step": 6470
|
| 5834 |
+
},
|
| 5835 |
+
{
|
| 5836 |
+
"epoch": 1.305661898045537,
|
| 5837 |
+
"grad_norm": 10.75,
|
| 5838 |
+
"learning_rate": 1.1296930619920748e-05,
|
| 5839 |
+
"loss": 0.8265,
|
| 5840 |
+
"mean_token_accuracy": 0.7950271546840668,
|
| 5841 |
+
"num_tokens": 7184403.0,
|
| 5842 |
+
"step": 6480
|
| 5843 |
+
},
|
| 5844 |
+
{
|
| 5845 |
+
"epoch": 1.307676808382027,
|
| 5846 |
+
"grad_norm": 14.9375,
|
| 5847 |
+
"learning_rate": 1.1283497884344149e-05,
|
| 5848 |
+
"loss": 0.7715,
|
| 5849 |
+
"mean_token_accuracy": 0.8035805761814118,
|
| 5850 |
+
"num_tokens": 7195550.0,
|
| 5851 |
+
"step": 6490
|
| 5852 |
+
},
|
| 5853 |
+
{
|
| 5854 |
+
"epoch": 1.309691718718517,
|
| 5855 |
+
"grad_norm": 13.8125,
|
| 5856 |
+
"learning_rate": 1.1270065148767547e-05,
|
| 5857 |
+
"loss": 0.8843,
|
| 5858 |
+
"mean_token_accuracy": 0.7886571526527405,
|
| 5859 |
+
"num_tokens": 7207013.0,
|
| 5860 |
+
"step": 6500
|
| 5861 |
}
|
| 5862 |
],
|
| 5863 |
"logging_steps": 10,
|
|
|
|
| 5877 |
"attributes": {}
|
| 5878 |
}
|
| 5879 |
},
|
| 5880 |
+
"total_flos": 8718380552103936.0,
|
| 5881 |
"train_batch_size": 8,
|
| 5882 |
"trial_name": null,
|
| 5883 |
"trial_params": null
|